Polyculture of Nile tilapia, Oreochromis niloticus, confined in cages suspended in prawn, Macrobrachium rosenbergii, ponds has been shown to reduce phytoplankton densities, pH levels, and increase prawn production when compared to prawn monoculture ponds. However, as filter feeders, tilapia grazing might be more efficient if allowed free access to all portions of the water column, positively impacting both phytoplankton control and tilapia growth. This study was designed to compare the effects of confined and unconfined tilapia, in polyculture with freshwater prawns, on prawn growth, tilapia growth, algae populations, and water quality. Juvenile prawns were stocked into each of nine, 0.04 ha ponds as 60 d nursed juveniles (0.8 ± 0.3 g) at 62,000/ha. Three control ponds contained only prawns (MONO). In three other ponds, monosex (male) Nile tilapia (89.2 ± 23.6 g) were stocked unconfined into three ponds at 4400/ha (POLY‐UNC). In three additional ponds, the same size and number of tilapia were stocked but confined in two, 1 m3 cages at 100 fish/cage (POLY‐CON). Prawns were fed a sinking pellet (28% protein) twice daily at a standardized rate for 114 d. Tilapia were fed a floating pellet (32% protein) twice daily to apparent satiation for 106 d. In the POLY‐UNC treatment, average prawn harvest weight (26 g) and prawn production (1625 kg/ha) were significantly lower (P < 0.05) and prawn feed conversion ratio (3.0) was significantly higher (P < 0.05) than the other two treatments. However, there were no significant differences (P > 0.05) between the MONO and POLY‐CON treatments in terms of prawn harvest weight, production and feed conversion ratio with combined averages of 38 g, 2465 kg/ha, and 1.9, respectively. Average harvest weight and production of adult tilapia were not significantly different (P > 0.05) in the POLY‐CON and POLY‐UNC treatments averaging 485 g and 2293 kg/ha, respectively. Tilapia in the POLY‐CON treatment had a significantly higher (P < 0.05) survival rate (99.7%) and feed conversion ratio (1.5) than POLY‐UNC (90.3 and 0.8, respectively). There were no consistent trends in treatment differences (P > 0.05) among water quality variables or phytoplankton populations. Lower prawn production in the POLY‐UNC is likely due to competition for food with the large number of tilapia juveniles. Despite the use of Genetically Modified Tilapia (GMT) monosex males, several cohorts of juvenile tilapia were produced in each of the POLY‐UNC ponds, resulting in over 2500 kg/ha of juveniles in the POLY‐UNC treatment. No wild spawn tilapia juveniles were found in the POLY‐CON ponds. In summary, confinement of tilapia in cages appears preferable when tilapia are polycultured with freshwater prawn.